Most of the existing research on child and adolescent mental health care and its effects is derived from controlled experiments involving conditions quite unlike those in most clinics. To complement that experimental data base, the proposed research will focus on outpatient treatment of children and adolescents (here referred to as children) in community clinics. Children aged 8-15 (initial N=545) will be assessed at clinic intake using established measures of adjustment, diagnosis, and parent-child communication. The measures will be repeated at 6 months, l year, and 2 years after intake, for (ad-children who continue for a full course of treatment, (b) those who start treatment but drop out prematurely, and (c) those who drop out immediately after intake. The longitudinal data thus generated will be used to test predictions associated with a multi-factor model of child treatment outcome. This model is applied to the individualized, family-oriented, outpatient treatment provided in five participating nonprofit child mental health clinics in a representative target zone of Los Angeles County. In the model, child outcomes are hypothesized to differ as a function of gender (with greater treatment benefits for girls than boys), ethnicity (with Latino children benefiting more than others), parent-child communication style (with treatment effects influenced favorably by positive styles and adversely by negative styles), and parent pathology (with treatment mitigating the impact of parent pathology on child outcome). Specific aims of the study are (a) to identify groups that are being well-served by current outpatient treatment and other groups for whom services may need to be modified or improved, and (b) to generate suggestions for enhancing the effects of community-clinic-based outpatient treatment. A broad, long-term objective is to complement our field's large base of controlled, experimental evidence with information on outpatient child treatment in practicing community clinics.